This paper considers the ultimate impact of fundamental physical limit
ations-notably, speed of light and device size-on parallel computing m
achines. Although we fully expect an innovative and very gradual evolu
tion to the limiting situation, we take here the provocative view of e
xploring the consequences of the accomplished attainment of the physic
al bounds. The main result is that scalability holds only for neighbor
ly interconnections, such as the square mesh, of bounded-size synchron
ous modules, presumably of the area-universal type. We also discuss th
e ultimate infeasibility of latency hiding, the violation of intuitive
maximal speedups, and the emerging novel processor-time tradeoffs. (C
) 1995 Academic Press, Inc.